I have two identical cooling towers on identical frequency drives. Tower 1 trips periodically. It gives a DC bus overvoltage fault code. At 12 Hz, Tower 1 motor draws 19 amps and 103V. Tower 2 motor draws 9 amps at 41V. MCE testing shows nearly identical results for each motor. The drive technician concluded that both drives operate properly but could not explain the voltage/current difference. Here is the kicker: When operated across the line, Tower 1 drew 16 amps and Tower 2 drew 36 amps. So, is the problem in the motor or in the drive? More detailed information is attached.

Cooling_Tower_Motor_Comparison.doc (69 KB, 32 downloads)

What HP motors are we talking about?

My first thought would be drive problem. However, the idea you have of running #2 off #1 drive would prove/disprove that.

D

The motors are 30 HP. 480V

RRS-Don't overlook the important detail of the currents on across the line operation. I would have suspected the drive until then, too.

Usually on cooling tower fans when the drive is set to run at very low frequencies the drive reacts much faster than the mechanical energy of the fan. The drive tends to overcorrect to the point that it creates an oscillation which can eventually trip the DC bus with an over current fault. When I saw the problem at lower frequencies, the drive ran fine at higher frequencies.

You have a very short sample time on your meter reading. On the fan that trips can you take a longer sample or multiple samples to see if they change?

The DC bus will pump itself up and down as it fights against the mechanical energy of the fan at most VFD default settings. The manufacturers pick default settings based on the majority of applications for which the drive will service, e.g. pumps, conveyors, etc. The remedy is to dampen the reaction time of the drive to slow its over-correction against the mechanical energy of the fan.

Across the line starts should be multiples higher than your rated FLA. Are you sure you have that sample at the highest starting current?

That’s all I can hazard a guess at for you. What kind of drives and how old are they?

Wally: I agree, the instrument that I used was crude. I can return with a power analyzer and properly monitor the drive. It is a Yaskawa drive. I would guess it is less than 5 years old. 100 amp.

Would you have any idea why the drive that is failing is producing 103V and 19 amps, while the other is running at 41V and 9 amps. Both at 12 Hz and both running identical motors at the same speed?

quote:

Originally posted by Testtech:
RRS-Don't overlook the important detail of the currents on across the line operation. I would have suspected the drive until then, too.

Yes, but with all things being equal on the cooling tower (you didn't change blade pitch) then at full speed, if the system is designed close, then you should have full current, or close thereof. Motor 1 isn't close. Motor 2 is.
At 12Hz motor 2 is close to what we would expect if we assume a linear rise in current (7.5 amps calculated vs. 9 amps actual).

Across the line I would expect higher current on a cooling tower.

D

This message has been edited. Last edited by: RRS_Dave,

Testtech,
The Yaskawa looks like the same drive that we saw the behavior I described previously. I picked the P7 as a guess for the drives you're running. Right up front, this may or may not be the problem, but I’d check the following parameters to see what their settings are.

For the parameter C2 S-Curve Acc you might try selecting C2-01 S-Curve Characteristic at Accel Start or equivalent. The slower S-Curve response along with a long acceleration time should dampen the drive’s correction factor for a cooling tower fan application.

When we saw the problem standing at the drive, the drive simply tripped out, but when we went out to the fan to see what it was doing, it was bucking back and forth to the point one of the guys thought the gearbox was bad before it tripped out. Turns out the drive was reacting too fast to control the high inertia load of the fan at low frequency set points.

Thanks for the input Wally.

I am encouraging the facility to swap the motor wiring to compare drive behavior. I did some calculations using the inductance measures provided by the MCE test. These calculations suggest a significant increase in impedance in the motor whose drive is tripping. I am wondering if this could account for high drive volt/amps at 12 Hz., as well as the tripping?

I am not a drive man, so please, bear with me.
However, I believe, there is a significant difference in the drives setting. If you need constant torque, you have to keep the ratio of the voltage to frequency constant. For 12 Hz it would be 480Volts*12Hz/60 Hz=96 Volts. It is almost exactly what the motor 1 sees. The second drive is programmed to do something significantly different. It does not keep the ratio of the voltage to frequency constant.
It does not do that, because you do not need constant torque to run a fan. The fan torque decreases sharply with the speed (something like cube of the speed). Hence, the 41 volts will easily run it at 12Hz. The 16 amps you see at 103 volts is probably mainly the magnetizing current (just guessing).
I cannot explain the currents across the line. The motor 1 with 36 amps would be fully loaded and pull the 30 hp. Motor #2 is not fully loaded. Only the watt measurement will tell you, if the loads are really identical.
Do not get carried away by the differences in impedances measured at low voltages. Even if there is some difference (and you say in the first posting that there is almost none), the impedance of the motor plays a very little role when the motor runs at full speed. It can be shown on the equivalent diagram, that what governs the current is NOT the motor impedance, but the load, or in other words resistance of the rotor divided by slip (R2/s). Compared to that, the motor impedance (as measured by low voltage tester) is almost negligible.
jank

We saw this problem on a cooling tower a few years ago. My tech solved it by changing parameters. Contact me and I can put you in touch with him. I am new to this site and blogging.leo@dreisilker.com

jank-that was a very astute observation. I have asked the facilities person to check drive settings to determine if the drive is set to maintain constant torque.

I have received a report back on this motor. They found the motor was wired backwards (don't know if that was primary or secondary). They wired the motor correctly. They believe the problem is corrected.

I will have to store this one way for the next time I see such symptoms.

Thanks for all of your comments.